Quantifying and modeling fabric surface roughness discrepancy: Consistency between physical and digital textures in online shopping

Abstract

In digital textile commerce, the absence of tactile interaction limits consumers’ ability to perceive fabric properties, often leading to mismatched expectations and product dissatisfaction. This study aimed to quantify the perceptual discrepancy in surface roughness (RDP) between physical fabrics and their digital representations and to identify structural parameters that influence this discrepancy. Plain- and twill-woven fabric specimens were prepared with varying densities, weights, and thicknesses. Surface roughness of physical fabrics was measured using atomic force microscopy (AFM), while digital roughness values were extracted from scanned images using ImageJ. Statistical analyses, including correlation and regression modeling, were applied to identify key predictors of RDP. Results showed that in plain-woven fabrics, lower weft density and higher warp density under fixed fabric weight conditions yielded the lowest RDP values, whereas in twill-woven fabrics, the perceptual gap was minimized at higher fabric weights and lower weft densities. These findings provide practical insights for improving visual–tactile alignment in virtual textile presentation and can inform fabric structural design for enhanced accuracy in online representation. Future research may explore nonlinear modeling and multisensory feedback systems to further reduce perceptual discrepancies.